Zobrazeno 1 - 10
of 523
pro vyhledávání: '"Huang, Zeyu"'
Reinforcement Learning from Human Feedback aligns the outputs of Large Language Models with human values and preferences. Central to this process is the reward model (RM), which translates human feedback into training signals for optimising LLM behav
Externí odkaz:
http://arxiv.org/abs/2409.17407
The scaling of large language models (LLMs) has revolutionized their capabilities in various tasks, yet this growth must be matched with efficient computational strategies. The Mixture-of-Experts (MoE) architecture stands out for its ability to scale
Externí odkaz:
http://arxiv.org/abs/2408.06793
Interdisciplinary studies often require researchers to explore literature in diverse branches of knowledge. Yet, navigating through the highly scattered knowledge from unfamiliar disciplines poses a significant challenge. In this paper, we introduce
Externí odkaz:
http://arxiv.org/abs/2408.00447
In this paper, we introduce a novel method called FRI-Net for 2D floorplan reconstruction from 3D point cloud. Existing methods typically rely on corner regression or box regression, which lack consideration for the global shapes of rooms. To address
Externí odkaz:
http://arxiv.org/abs/2407.10687
Convolutional neural networks (CNNs), one of the key architectures of deep learning models, have achieved superior performance on many machine learning tasks such as image classification, video recognition, and power systems. Despite their success, C
Externí odkaz:
http://arxiv.org/abs/2407.11031
Large Language Models (LLMs) excel in fluency but risk producing inaccurate content, called "hallucinations." This paper outlines a standardized process for categorizing fine-grained hallucination types and proposes an innovative framework--the Progr
Externí odkaz:
http://arxiv.org/abs/2407.00488
Mixture-of-experts (MoE) is gaining increasing attention due to its unique properties and remarkable performance, especially for language tasks. By sparsely activating a subset of parameters for each token, MoE architecture could increase the model s
Externí odkaz:
http://arxiv.org/abs/2406.18219
Autor:
Du, Wenyu, Cheng, Shuang, Luo, Tongxu, Qiu, Zihan, Huang, Zeyu, Cheung, Ka Chun, Cheng, Reynold, Fu, Jie
Language models (LMs) exhibit impressive performance and generalization capabilities. However, LMs struggle with the persistent challenge of catastrophic forgetting, which undermines their long-term sustainability in continual learning (CL). Existing
Externí odkaz:
http://arxiv.org/abs/2406.17245
Autor:
Du, Wenyu, Luo, Tongxu, Qiu, Zihan, Huang, Zeyu, Shen, Yikang, Cheng, Reynold, Guo, Yike, Fu, Jie
LLMs are computationally expensive to pre-train due to their large scale. Model growth emerges as a promising approach by leveraging smaller models to accelerate the training of larger ones. However, the viability of these model growth methods in eff
Externí odkaz:
http://arxiv.org/abs/2405.15319
In this paper, we introduce a new method for the task of interaction transfer. Given an example interaction between a source object and an agent, our method can automatically infer both surface and spatial relationships for the agent and target objec
Externí odkaz:
http://arxiv.org/abs/2405.03221